• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.1
• /
• pp.77-88
• /
• 2022

With the development and popularization of drone manufacturing technology, the drone market, which was mainly focused on industry, agriculture, and military, is also showing great growth in individual and commercial markets. Among them, DJI has a high share in the personal and commercial drone market, and accordingly, forensic analysis of DJI drones is drawing attention. In particular, when stealing and analyzing drones used in criminal acts, a technology to interpret flight logs recording drone flight paths and hardware information is needed, which inevitably applies drone models due to differences in decryption methods. Therefore, when an unidentified drone is acquired from the perspective of a digital forensic investigator, a clear classification of a drone model to which analysis can be applied is required. This paper proposes a method of extracting and analyzing artifacts of DJI drones through forensics, and analyzes media data analysis and flight log analysis results and decryption methods for three drone models with different release years of DJI. Finally, drones in the DJI product line are classified according to whether the commercialized DJI drone flight log is decrypted.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.333-334
• /
• 2018

This is a study on controlling a drone (DJI Mavic Air) with simple hand gesture using Leap Motion controller. Four component involve are MacBook, Leap Motion controller, Android device, and DJI Mavic Air. All of this component are connected through USB, Bluetooth, and Wi-Fi technology. The studies main purpose are to show that by controlling a drone through Leap Motion, drone amateur user can easily learn how to control a drone, and because of longer drone control range can be archived things such as search and rescue mission will be possible.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.609-619
• /
• 2023

Recently, crimes using drones, one of the IoT industries have been continuously reported. In particular, drones are characterized by easy access and free movement, so they are used for various crimes such as transporting explosives, transporting drugs, and illegal recording. In order to analyze and investigate these criminal acts, drone forensic research is highly emphasized. Media data, PII, and flight records are digital forensic artifacts that can be acquired from drones, in particluar flight records are important artifacts since they can be used to trace drone activities. Therefore, in this paper, the characteristics of the deleted flight record files of DJI drones are presented and verified using the Phantom3, Phantom4 andMini2 models, two drones with differences in characteristics. Additionally, the recovery level is analyzed using the flight record file characteristics, and lastly, drones with the capacity to recover flight records for each drone model and drone models without it are classified.

• Journal of Aerospace System Engineering
• /
• v.14 no.3
• /
• pp.10-16
• /
• 2020

We propose a capture mechanism based on the principles of fishing nets that can be mounted on the drone using cable. The gripper mechanism, mainly proposed for the drone is heavy, and is limited to catch standardized objects. In contrast, the proposed capture device in this paper is light, flexible, and can capture various types of objects from a long distance. The theoretical relationships between cables and mechanisms were analyzed. Finally, the capture device was designed and manufactured to be installed in the drone (DJI S900) to conduct capturing experiments for various objects and verify the validity.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.117-119
• /
• 2021

With the development of the industry, the use of drones in specific mission flight is being actively studied. These drones fly a specified path and perform repetitive tasks. if the drone system will detect objects in real time, the performance of these mission flight will increase. In this paper, we implement object detection system and mount GPU acceleration to maximize the efficiency of limited device resources with drone video using Tensorflow Lite which enables in-device inference from a mobile device and Mobile SDK of DJI, a drone manufacture. For performance comparison, the average processing time per frame was measured when object detection was performed using only the CPU and when object detection was performed using the CPU and GPU at the same time.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.1
• /
• pp.143-149
• /
• 2018

This paper presents an implementation of a streaming system which can forward live video stream to multiple users from a Phantom4, which is a drone made by DJI. We constructed the streaming server on Raspberry Pi 3 board for high mobility. Also We implemented the system so that the video stream can be played on any devices if the HTML5 standard web browser is utilized. We compiled C codes of FFmpeg open sources and installed in the Raspberry Pi3 as the streaming server and developed a Java application to execute as the integrated server that controls the other softwares on the streaming server. Also we developed an Android application which receives the live video stream from the drone and sends the streaming server continuously. The implemented system in this paper can successfully stream the live video on 24 frames per second at the resolution of 148x112 in considering the low hardware throughput of the streaming server.

• Journal of Aerospace System Engineering
• /
• v.17 no.4
• /
• pp.133-143
• /
• 2023

As drones (also known as UAV) become popular with advanced information and communication technology (ICT), they have been utilized for various fields (agriculture, architecture, and so on). However, malicious attackers with advanced drones may pose a threat to critical national infrastructures. Thus, anti-drone systems have been developed to respond to drone threats. In particular, remote identification data (R-ID)-based UAV detection and identification systems that detect and identify illegal drones with R-ID broadcasted by drones have been developed, and are widely employed worldwide. However, this R-ID-based UAV detection/identification system is vulnerable to security due to wireless broadcast characteristics. In this paper, we analyze the security vulnerabilities of DJI Aeroscope, a representative example of the R-ID-based UAV detection and identification system, and propose a replay-attack-based neutralization method using the analyzed vulnerabilities. To validate the proposed method, it is implemented as a software program, and verified against four types of attacks in real test environments. The results demonstrate that the proposed neutralization method is an effective neutralization method for R-ID-based UAV detection and identification systems.

• Information and Communications Magazine
• /
• v.33 no.2
• /
• pp.80-85
• /
• 2016

군사용 목적으로 개발된 드론(drone)이 오픈소스로 시장을 형성하면서 급성장하여 우체국의 택배 시스템에 적용을 한다는 발표[10]를 보면 국가 신성장 동력으로 발전 가능성에 대하여 기대하고 있는 실정이다. 언론과 산업계에서 드론이 모든 것을 대체할 것이라고 인식되면서 기술의 한계점은 흐려지고 있다. 세계 시장에서는 중국의 DJI를 선두로 급속히 성장하고 있지만 규제 목소리 또한 만만치 않은 실정이다. 플랫폼 기업인 구글이나 페이스북 아마존 등에서는 드론을 적용[3]하기 시작하였지만 넘어야 할 산은 많은 것이 현실이다. 이러한 드론 기술의 수준과 해결 과제를 살펴보고 드론이란 무엇이며 국내외 동향을 살펴본다.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.319-327
• /
• 2022

Robots for construction sites, although not deeply widespread, are finding applications in the duties of project monitoring, material movement, documentation, security, and simple repetitive construction-related tasks. A significant shortcoming in the use of robots is the complexity involved in programming and re-programming an automation routine. Robotic programming is not an expected skill set of the traditional construction industry professional. Therefore, this research seeks to deliver a low-cost approach toward re-programming that does not involve a programmer's skill set. The researchers in this study examined an approach toward programming a terrestrial-based drone so that it follows a taped path. By doing so, if an alternative path is required, programmers would not be needed to re-program any part of the automated routine. Changing the path of the drone simply requires removing the tape and placing a different path - ideally simplifying the process and quickly allowing practitioners to implement a new automated routine. Python programming scripts were used with a DJI Robomaster EP Core drone, and a terrain navigation assessment was conducted. The study examined the pass/fail rates for a series of trial run over different terrains. The analysis of this data along with video recording for each trial run allowed the researchers to conclude that the accuracy of the tape follow technique was predictable on each of the terrain surfaces. The accuracy and predictability inform a non-coding construction practitioner of the optimal placement of the taped path. This paper further presents limitations and suggestions for some possible extended research options for this study.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.3
• /
• pp.293-302
• /
• 2023

This study examines and proposes standardization approaches to address the heterogeneous issues of metadata in drone imagery within the agricultural sector. Image metadata comes in various formats depending on different camera manufacturers, with most utilizing EXIF and XMP. The metadata of cameras used in fixed-wing and rotary-wing platforms, along with the metadata requirements in image alignment software, were analyzed for sensors like DJI XT2, MicaSense RedEdge-M, and Sentera Double4K. In the agricultural domain, multispectral imagery is crucial for vegetation analysis, making the provision of such imagery essential. Based on Pix4D SW, a comparative analysis of metadata attributes was performed, and necessary elements were compiled and presented as a proposed standardization (draft) in the form of tag information.